It is often assumed that deep-sea ecosystems are shielded from the effects of climate change at the surface. On the contrary, such ecosystems are likely to be particularly sensitive to changing oceanic conditions. For one thing, many are energetically dependent on organically rich particles, which are produced in surface waters before sinking to the sea floor as 'marine snow'. Furthermore, because many deep ecosystems have experienced relatively constant conditions for millennia, even small perturbations of the physical and chemical environment could destabilize them. Many of the species supported by these ecosystems have long life spans and generation times, meaning that their capacity to adapt quickly enough to keep pace with environmental change may be limited. So what, if anything, can be done to protect them?

This was the topic of a meeting of stakeholders held in Hobart, Tasmania, the outcomes of which are discussed by Ronald Thresher and colleagues in this issue (page 635). The focus of the workshop was the deep-sea coral communities of the Huon Commonwealth Marine Reserve, off southeast Australia. These reef systems were established before the peak of the last Ice Age, and now have protected status. However, ocean acidification is likely to reduce the ability of taxa such as reef-building corals to calcify their skeletons. Without management intervention, these cold-water reef systems may well be seriously degraded or even lost within decades, a dire situation indeed.

To prevent this, the workshop recommended that short-term priority should be given to identifying and protecting sites in the world's oceans that are, or could become, refugia areas. But what other actions can be taken to protect deep-sea reefs? The Huon reefs are located on seamounts at depths of 1,000 meters or more — representing an important challenge in itself. Other cold-water reef systems are found at even greater depths elsewhere around the world; making it hard to observe and monitor them, let alone manage them through direct intervention. The workshop considered 17 options, some more desirable — or at least less undesirable — than others, and ranging in probable cost, practicality, and lead times for implementation.

Engineering-based solutions might include providing concrete structures as homes for species that normally live around live coral, or dumping bags of lime to improve conditions for coral growth. Biological interventions could potentially involve strategic translocation of colonies to or from refugia areas, or even genetically engineering coral species to make them more tolerant of warmer and more acidic conditions. Changes to regulatory frameworks to expand the size of legislatively protected areas might also be beneficial.

The truth of the matter is that deep-sea reef systems the world over face a perfect storm of threats and all attempts to save them may, unfortunately, prove futile.